Ocean mesoscale vortex three-dimensional temperature field inversion method, device, equipment and medium

Abstract

The invention discloses an ocean mesoscale vortex three-dimensional temperature field inversion method, device and equipment and a medium, and relates to the field of ocean observation and data inversion, and the method comprises the steps: obtaining multi-source ocean observation data; based on the multi-source ocean observation data, performing mesoscale vortex identification, and determining a mesoscale vortex three-dimensional structure; constructing a mesoscale eddy three-dimensional temperature field inversion model by adopting a regression analysis method according to the mesoscale eddy three-dimensional structure based on the multi-source ocean observation data; solving a regression coefficient in the mesoscale eddy three-dimensional temperature field inversion model by adopting a hierarchical least square method; and according to the regression coefficient, the mesoscale eddy three-dimensional temperature field inversion model is adopted to invert a mesoscale eddy three-dimensional temperature field, and dynamic constraints are embedded in the inversion process so as to obtain the temperature of each standard layer of the mesoscale eddy. According to the invention, the precision and reliability of the temperature inversion result of each standard layer of the mesoscale vortex are improved.

Description

Technical Field

[0001] This application relates to the field of marine observation and data inversion, and in particular to a method, apparatus, equipment and medium for inverting the three-dimensional temperature field of ocean mesoscale eddies. Background Technology

[0002] Mesoscale eddies in the ocean are key carriers of energy transfer and heat transport, and their three-dimensional temperature field characteristics are crucial for ocean thermal balance, climate prediction, and marine resource development. Current methods for obtaining their temperature fields have limitations: in-situ observations offer high precision but lack spatial coverage and temporal sampling, leading to observational blind spots; satellite remote sensing allows for large-scale observations but can only acquire surface ocean temperature, failing to obtain three-dimensional structure. Currently, the inversion of the three-dimensional structure of mesoscale eddies is affected by the spatiotemporal resolution and data accuracy of satellite remote sensing, making it impossible to accurately characterize the three-dimensional structure of mesoscale eddies.

[0003] In summary, existing technologies are insufficient to achieve high-precision and high-reliability inversion, and there is an urgent need for a method that can fully integrate multi-source data, adapt to the characteristics of ocean stratification, and conform to the laws of dynamics. Summary of the Invention

[0004] The purpose of this application is to provide a method, apparatus, equipment and medium for inverting the three-dimensional temperature field of ocean mesoscale eddies, which can improve the accuracy and reliability of the temperature inversion results of each standard layer of mesoscale eddies.

[0005] To achieve the above objectives, this application provides the following solution: Firstly, this application provides a method for inverting the three-dimensional temperature field of ocean mesoscale eddies, including: Acquire multi-source ocean observation data; Based on the multi-source ocean observation data, mesoscale eddies are identified, and their three-dimensional structures are determined. Based on the multi-source ocean observation data, and according to the three-dimensional structure of the mesoscale eddy, a three-dimensional temperature field inversion model of the mesoscale eddy is constructed using regression analysis. The regression coefficients in the mesoscale eddy three-dimensional temperature field inversion model are solved using the hierarchical least squares method. Based on the regression coefficients, the mesoscale eddy three-dimensional temperature field inversion model is used to invert the mesoscale eddy three-dimensional temperature field. At the same time, dynamic constraints are embedded in the inversion process to obtain the temperature of each standard layer of the mesoscale eddy.

[0006] Secondly, this application provides a three-dimensional temperature field inversion device for ocean mesoscale eddies, comprising: The data acquisition module is used to acquire multi-source ocean observation data; The mesoscale eddy identification module is used to identify mesoscale eddies based on the multi-source ocean observation data and determine the three-dimensional structure of the mesoscale eddies. The model building module is used to construct a three-dimensional temperature field inversion model of a mesoscale eddy based on the multi-source ocean observation data and the three-dimensional structure of the mesoscale eddy using regression analysis. The coefficient solving module is used to solve the regression coefficients in the mesoscale eddy three-dimensional temperature field inversion model using the hierarchical least squares method. The temperature inversion module is used to invert the three-dimensional temperature field of the mesoscale eddy using the mesoscale eddy three-dimensional temperature field inversion model based on the regression coefficients, and to embed dynamic constraints during the inversion process to obtain the temperature of each standard layer of the mesoscale eddy.

[0007] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described method for inverting the three-dimensional temperature field of ocean mesoscale eddies.

[0008] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for inverting the three-dimensional temperature field of ocean mesoscale eddies.

[0009] According to the specific embodiments provided in this application, this application has the following technical effects: This application provides a method, apparatus, equipment, and medium for inverting the three-dimensional temperature field of ocean mesoscale eddies. By acquiring multi-source ocean observation data, it ensures the comprehensiveness and reliability of the basic data required for inversion, providing high-quality data support for subsequent three-dimensional temperature field inversion of mesoscale eddies, avoiding information bias caused by the limitation of a single data source. First, it identifies mesoscale eddies and determines their three-dimensional structure based on multi-source data, and then uses regression analysis to model them, so that the three-dimensional temperature field inversion model of mesoscale eddies can fit the actual morphology of the eddies, improving the model's relevance and rationality, reducing the error of blind modeling that deviates from the actual structure, and using the hierarchical least squares method to solve the regression coefficients, which can adapt to the characteristic differences of different ocean layers, improve the accuracy of coefficient calculation, and enhance the model's ability to fit different ocean layers. By embedding dynamic constraints in the inversion, it ensures that the inversion results conform to the laws of ocean dynamics, effectively avoids unreasonable temperature values, and ultimately improves the reliability of the temperature inversion results of each standard layer of mesoscale eddies. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1This is an application environment diagram of a three-dimensional temperature field inversion method for ocean mesoscale eddies in one embodiment of this application.

[0012] Figure 2 This is a flowchart illustrating a method for inverting the three-dimensional temperature field of a mesoscale eddy in the ocean, provided as an embodiment of this application.

[0013] Figure 3 This is a schematic diagram of the functional modules of a three-dimensional temperature field inversion device for ocean mesoscale eddies provided in an embodiment of this application.

[0014] Figure 4 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0015] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0016] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0017] The three-dimensional temperature field inversion method for ocean mesoscale eddies provided in this application can be applied to, for example... Figure 1 In the application environment shown, terminal 101 communicates with server 102 via a network. A data storage system can store the data that server 102 needs to process. The data storage system can be set up independently, integrated into server 102, or placed in the cloud or on another server. Terminal 101 can send multi-source ocean observation data to server 102. After receiving the multi-source ocean observation data, server 102 performs mesoscale eddy identification and constructs a three-dimensional temperature field inversion model for the mesoscale eddy to invert the three-dimensional temperature field of the mesoscale eddy. Server 102 can feed back the obtained temperatures of each standard layer of the mesoscale eddy to terminal 101. Furthermore, in some embodiments, the three-dimensional temperature field inversion method for ocean mesoscale eddies can also be implemented independently by server 102 or terminal 101.

[0018] The terminal 101 can be, but is not limited to, various desktop computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, and smart in-vehicle devices. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted devices. The server 102 can be implemented using a standalone server or a server cluster composed of multiple servers, or it can be a cloud server.

[0019] In one exemplary embodiment, such as Figure 2 As shown, a method for inverting the three-dimensional temperature field of ocean mesoscale eddies is provided. This method is executed by computer equipment, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method is applied to... Figure 1 Taking server 102 as an example, the explanation includes the following steps 201 to 205.

[0020] Step 201: Acquire multi-source ocean observation data.

[0021] The multi-source ocean observation data includes buoy temperature and salinity datasets, satellite altimeter data, sea surface temperature reanalysis data, ocean climate datasets, and drifting air-sea interface buoy observation data. Each data point is described in detail below.

[0022] (1) The buoy temperature and salinity dataset adopts the Array for Real-time Geostrophic Oceanography (Argo) dataset, which has 58 vertical layers. The Argo dataset underwent real-time quality checks and post-processing quality control. The original dataset includes approximately 600,000 temperature and salinity profiles covering the globe, as well as the time and location information of the profile observations, with the profiles extending from a depth of 2000m to the sea surface.

[0023] The filtering criteria for the Argo dataset in this application are as follows.

[0024] 1) The shallowest data in the Argo profile is located between the sea surface and 10 dbar, while the deepest data point corresponds to a depth greater than 1000 dbar.

[0025] 2) The depth interval between two consecutive data points meets the following filtering criteria: the data interval between 0-100 dbar does not exceed 25 dbar; the data interval between 100-300 dbar does not exceed 50 dbar; and the data interval between 300-1000 dbar does not exceed 100 dbar.

[0026] (2) In the AVISO project, the global gridded sea surface height products include two types: Sea Level Anomaly (SLA) and Absolute Dynamic Topography (ADT). SLA is approximately obtained by subtracting the mean sea surface height (MSSH) over a certain period from the sea surface height (SSH). MSSH is usually the average sea surface height over 7 years (1993-1999). ADT is equal to SLA plus mean dynamic topography (MDT). Here, MDT is calculated from 4.5 years of Gravity Recovery and Climate Experiment (GRACE) data, 15 years of altimeter data, and field data. Subtracting an average value from SSH or ADT aims to weaken the influence of mean currents and highlight mesoscale signals. Therefore, SLA is more suitable for studying mesoscale eddies, while ADT is often used for studying general ocean circulation.

[0027] SLA data is available in both daily and weekly formats, incorporating data from multiple orbiting satellites including T / P, Jason-1, Jason-2, and Geosat Follow-On. This application uses daily SLA data.

[0028] Furthermore, it should be noted that although AVISO has corrected its altimeter products to eliminate errors caused by factors such as tidal channel surface pressure, this influence still exists in shallow water areas. Previous studies would have removed data from depths shallower than a certain isobath (e.g., 200m), but given the specific nature of this application (the research depth involves depths shallower than 200m), completely discarding data from shallow water areas would inevitably affect the identification of eddies near that isobath. Therefore, this application has removed data from land and large islands, but retained data from small islands.

[0029] (3) The sea surface temperature reanalysis data is the Optimum Interpolation Sea Surface Temperature (OISST) dataset. OISST is an improved version of the Comprehensive Ocean-Atmosphere Data Set (COADS) that utilizes data from different observation platforms, such as satellites, buoys, and ship surveys. It employs rigorous quality control of satellite observation data, including adjustments for deviations between satellite observations and ship surveys (relative to reference buoys), compensation for differences between platforms and sensor-induced biases, and uses the Optimum Interpolation (OI) algorithm to interpolate buoy and ship data into the sea surface temperature gaps retrieved from remote sensing, thereby improving the spatial coverage and spatiotemporal resolution of the data. The OISST dataset begins in 1981 and has a horizontal resolution of 0.25° × 0.25°. This application uses global OISST data from 2016 to the present.

[0030] (4) Marine climate dataset.

[0031] Inversion of the three-dimensional temperature field structure of mesoscale eddies generally requires the use of outliers. To obtain these outliers, an average value needs to be removed from the Argo profile data. In practice, this average value can be replaced by climatological data. The climatological data used in this application is the WOA18 dataset. The WOA18 dataset provides objectively analyzed standard stratospheric temperature and salinity field data with a spatial horizontal resolution of 1°×1°. This application uses the annual average data of WOA18 as climatological temperature and salinity field data to calculate the temperature and salinity anomalies caused by eddies.

[0032] (5) The observation data of the drifting air-sea interface buoy (DrIB) can measure 11 different physical parameters, including air temperature, air pressure, wind direction, wind speed, relative humidity, sea surface temperature (20cm underwater), and wave parameters, up to 3m above the sea surface. The time interval of the SST observation data used in this application is 1 hour, and the observation accuracy is ±0.01℃.

[0033] Furthermore, the aforementioned multi-source ocean observation data were standardized. First, sea surface height anomaly (SSH, SLA) data provided by satellite altimeters, vertical temperature and salinity profile data from in-situ observations (Argo), and objective reanalysis data were integrated. Strict quality control was performed on the raw data, outliers were removed, and in-situ observations were interpolated vertically to the standard depth layer to ensure data consistency and reliability. The specific steps are as follows.

[0034] (1) Data source integration.

[0035] Satellite data: OISST, SLA, SSH, ADT.

[0036] Field observation data: Argo profile temperature data (0-200 dbar), DrIB water temperature data.

[0037] Background field data: Climatic temperature profile Tclim(z), source WOA18.

[0038] (2) Data quality control.

[0039] To improve the accuracy of the results, the above data underwent quality control and screening. Quality control mainly consisted of two steps: quality control of the observation layer and quality control of the standard layer.

[0040] Quality control of the observation layer includes the following steps.

[0041] 1) Regional test: The range of the study area (0-180°E, 0-60°N) and the range of the study time (2018) were selected.

[0042] 2) Duplicate depth data check: Filter out and delete data with the same depth in each temperature and salinity observation profile.

[0043] 3) Delete the data with large root mean square error and calculate the average value of the remaining data, which will be used as the final observation value for that depth.

[0044] 4) Stability test: Remove observation profiles that change too much in the vertical direction.

[0045] 5) Range control: Select a data range that matches the actual data range and remove NaN values.

[0046] 6) Verification of duplicate profiles: Since the historical observation data collected come from different databases, there may be multiple highly overlapping or even duplicate temperature and salinity observation profiles at the same location and time. Select the profile data with better quality.

[0047] 7) Outlier removal: Remove data whose absolute deviation is greater than 3 × interquartile range.

[0048] Quality control of the standard layer includes the following steps.

[0049] 1) A standard layer dataset was constructed by interpolating the temperature observation profiles after the aforementioned quality control. The dataset consists of 17 vertical layers: 0m, 5m, 10m, 15m, 20m, 25m, 30m, 35m, 40m, 45m, 50m, 75m, 100m, 125m, 150m, 175m, and 200m. The stability of this standard layer data was then verified to obtain a field-observed temperature-salinity profile dataset suitable for this application.

[0050] 2) Perform interpolation calculations for each depth layer to be interpolated. Extract the vortex water stratification data of the upper and lower layers of the interpolation layer. For example, for the first 40-meter water layer to be interpolated, extract the vortex water stratification data of the 35-meter and 50-meter layers.

[0051] 3) Calculate the centroid of the upper and lower vortex regions. Based on the distance between the interpolation layer and the depths of the upper and lower layers, calculate the position of the centroid of the interpolation layer.

[0052] Step 202: Based on the multi-source ocean observation data, identify mesoscale eddies and determine their three-dimensional structure.

[0053] In a specific application example, when performing the inversion of mesoscale eddies, it is first necessary to accurately identify the locations of the eddy core and eddy edge of the mesoscale eddy.

[0054] This application targets the sea surface region and, based on the aforementioned satellite altimeter data, employs the parametric method (Okubo Weiss, OW) to identify mesoscale eddies, thereby determining the three-dimensional structure of mesoscale eddies in the sea surface region.

[0055] The OW method starts from the physical properties of mesoscale eddies, identifying them through sea surface height (SSH) or sea surface temperature, and parameterizing the information of the physical field. When this parameter exceeds a set threshold, it is identified as a mesoscale eddy. This application uses sea surface height anomaly (SLA) and sea surface height (SSH) to identify eddies.

[0056] OW number is defined as: ;in, W For OW number, Sn For normal strain, , Ss For tangential strain, , w Relative vorticity . Let be the differential of the meridional component of the flow direction along the x-axis. Let be the differential of the meridional component of the flow direction along the y-axis. Let x be the differential of the latitudinal component of the flow direction along the x-axis. It is the differential of the latitudinal component of the flow direction in the y-axis direction.

[0057] Assume the flow field satisfies the geostrophic equilibrium relationship, i.e. =0. Therefore, the OW number can be simplified to: .

[0058] Geostrophic relationship calculated based on SLA , ,in, u The latitudinal component represents the flow direction. v The meridional component represents the direction of flow. g The acceleration due to gravity is taken as 9.8 m / s². 2 , f For Coriolis parameters, Ω is the angular rate of Earth's rotation. Latitude represents the effect of the Earth's rotation on flow, measured in units of 1 / s. The OW number can then be transformed into: ;in, The mixed second-order partial derivative of the sea level height; The second derivative (curvature) of sea level height in the eastward direction reflects the degree of curvature of the SLA contour lines in the eastward direction. Positive values ​​indicate upward bulging (local high pressure), while negative values ​​indicate downward denting (local low pressure). This is the second derivative of sea level height in the north direction, reflecting the curvature of the SLA contour lines in the north direction.

[0059] If |W| < 0.2σ in a certain region w It is then determined to be a mesoscale eddy, σ w This represents the standard deviation of the OW number for the region. Specifically, the flow field is divided into different types: W > 0.2σ. w Primarily tensile, W < -0.2σ w Primarily based on vorticity, the absolute value of W is ≤0.2σ. w This is the background flow field. W < -0.2σ w The region is considered the center of the vortex, and the SLA value determines whether it is a cyclonic vortex or an anticyclonic vortex.

[0060] For the subsurface region (5m-200m layer), based on the satellite altimeter data, the Winding-Angle (WA) method is used to identify mesoscale eddies in order to determine the three-dimensional structure of mesoscale eddies in the subsurface region.

[0061] The core logic of the WA method is that a vortex, as a closed circulation structure with "rotational" characteristics in the flow field, has a velocity vector field that satisfies specific geometric constraints (such as linear variation of the radial velocity component and consistency of rotation direction), and the stream function contour lines are closed. The WA method first filters vortex centers that meet the constraints, then traces the outermost closed streamlines centered on the center, ultimately determining the vortex boundary and type. The entire process does not rely on the calculation of physical parameters such as vorticity and divergence, but is based solely on the geometric characteristics of the flow field for identification. Simply put, it first calculates streamlines from the velocity field, then retains closed streamlines according to the Winding-Angle principle; streamlines clustered together are then classified as a vortex. The implementation of the WA method typically involves two steps: vortex center detection and vortex boundary delineation. Its input data includes SLA, SSH, and their corresponding latitude, longitude, and time data, and its output data includes the latitude and longitude of the vortex edge, the latitude and longitude of the vortex center, and the vortex radius.

[0062] (1) Vortex center detection: First, it is necessary to find points in the flow field that may become vortex "seeds". The WA method is used to achieve this by analyzing the behavior of the velocity vector around each grid point. The specific steps are as follows.

[0063] 1) Calculate the reference angle. For each point in the flow field... (Candidate point), check multiple neighboring points around it. The velocity vector from the candidate point. Point to adjacent points vector: , This is the radial vector. Calculate the velocity vector. With radial vector The angle between .

[0064] 2) Apply geometric constraints. Velocity direction constraint: velocity vector. It should be approximately perpendicular to the radial vector. This means Approaching 90° or 270° (i.e., tangential flow), rather than towards or away from the center (radial flow). Rotational consistency constraint: the velocity direction of all surrounding points should exhibit a systematic clockwise or counterclockwise rotational pattern.

[0065] 3) Filter candidate centers. Calculate the proportion of velocity vector points within a small region that satisfy the above constraints. If the proportion exceeds a set threshold, then that point is considered a candidate point. It is marked as a candidate vortex center.

[0066] (2) Delineation of vortex boundaries: After finding the candidate vortex center, the boundaries of the vortex are determined. The specific steps are as follows.

[0067] 1) Calculate the streamline integral and the winding angle. Starting from the identified vortex center, integrate the flow field in the reverse direction to generate a streamline. The winding angle is defined as the angle of cumulative change in the velocity vector direction of a particle moving around the vortex center on the streamline. Specifically, calculate the change in velocity direction point by point along the streamline and accumulate it to obtain the winding angle.

[0068] 2) Identifying Closed Streamlines. For a streamline to form a closed loop, its velocity vector must rotate a full circle. Therefore, the absolute value of the winding angle corresponding to a closed streamline must be close to 2π (360°). Specifically, examine the streamlines integrated from the center. When the absolute value of the winding angle of a streamline first reaches or exceeds 2π, it is determined that an outermost closed streamline has been found.

[0069] 3) Determine the final boundary. The region enclosed by the outermost closed streamlines is the boundary of the vortex. At the same time, determine the type of vortex (cyclonic or anticyclonic) based on the direction of streamline integration (clockwise or counterclockwise).

[0070] In addition, the vortex center can be determined by the local extrema of SSH within the grid range, and then the streamlines in its vicinity can be calculated. The outermost closed streamline is taken as the vortex edge, and SSH contour lines are used to replace the streamlines, thereby saving the time of calculating streamlines.

[0071] Step 203: Based on the multi-source ocean observation data and according to the three-dimensional structure of the mesoscale eddy, a three-dimensional temperature field inversion model of the mesoscale eddy is constructed using regression analysis.

[0072] In a specific application example, historical observation and reanalysis data are first integrated. By quantifying the statistical correlation between the dependent and independent variables, a predictive mathematical function is established and then extended to temperature field prediction. This application designs 17 standard layers (0m, 5m, 10m, 15m, 20m, 25m, 30m, 35m, 40m, 45m, 50m, 75m, 100m, 125m, 150m, 175m, 200m). Step 203 includes steps 31 to 36.

[0073] Step 31: Determine the sea surface temperature data (SST) based on the sea surface temperature reanalysis data, which reflects the intensity of vortex thermal forcing (cyclonic vortices correspond to cold anomalies, and anticyclonic vortices correspond to warm anomalies).

[0074] Step 32: Determine sea surface height anomaly data (SLA) based on the satellite altimeter data to characterize the potential energy change of the vortex dynamic core, which is directly related to the vortex's rotational speed and vertical scale.

[0075] Step 33: Based on the buoy temperature and salinity dataset (Argo temperature), the ocean climate dataset (WOA18), and the drifting air-sea interface buoy observation data (DrIB observed water temperature), determine the sea surface thermal anomaly data (SST′) and capture the asymmetric effects of temperature anomalies (such as the nonlinear response differences between cold and warm eddies).

[0076] Step 34: Based on the sea surface thermal anomaly data and the sea surface height anomaly data, determine the thermodynamic coupling data and quantify the thermodynamic-dynamic coupling effect (such as the synergistic influence of SST' and SLA on thermocline deformation). The thermodynamic coupling data is the product of the sea surface thermal anomaly data and the sea surface height anomaly data, i.e., SST' × SLA.

[0077] Step 35: Based on the sea level height anomaly data, determine the secondary modulation data of the vortex intensity vertical structure, characterizing the secondary modulation of the vertical structure by the vortex intensity. The secondary modulation data of the vortex intensity vertical structure is the square of the sea level height anomaly data.

[0078] Step 36: Using the sea surface height anomaly data, the sea surface thermal anomaly data, and the thermodynamic coupling data as independent variables, and the temperature of each standard layer of the mesoscale eddy as the dependent variable, a three-dimensional temperature field inversion model of the mesoscale eddy is constructed using regression analysis. This involves the SST in Step 31 and the temperature of each standard layer in Step 35. All of these are data used in the calculation process.

[0079] For each standard depth layer, this application establishes an independent mesoscale eddy three-dimensional temperature field inversion model: .

[0080] in, z For depth coordinates, from the sea surface ( z =0) indicates a downward trend with increasing units in meters. For depth Temperature at the standard floor level This is data on sea surface thermal anomalies. This is data on sea level height anomalies. For thermodynamic coupling data, For random error term, The efficiency of vertical transfer of sea surface thermal anomalies to deep water is modulated by mixed-layer convection (surface layer) and turbulent diffusion (deep layer). , , , For regression coefficients, The linear response coefficient representing the sea surface thermal anomaly characterizes the temperature change caused by the SLA through the displacement of the thermocline. The linear response coefficient representing the sea surface height anomaly characterizes the vertical displacement of the thermocline corresponding to a unit change in SLA (the magnitude decreases exponentially with depth). The nonlinear response coefficient representing sea surface thermal anomalies is used to quantify the asymmetry in the thermal response of cold / warm eddies, stemming from the differential response of the mixing layer depth to eddy polarity. The coefficient represents the thermodynamic coupling response, characterizes the thermodynamic-dynamic synergistic effect, and reflects the mathematical derivation of the modulation efficiency of the thermocline displacement by sea surface thermal anomalies.

[0081] Step 204 involves using the hierarchical least squares method to solve for the regression coefficients in the mesoscale eddy three-dimensional temperature field inversion model. In the hierarchical least squares solution process, an independent training set is constructed for each standard layer, with the objective defined as minimizing the mean square error between the predicted temperature and the observed value. The formulas for each regression coefficient are described below.

[0082] (1) Employing an exponential decay + asymptotic background model: .

[0083] in, For surface response intensity, , N The total number of observed samples used for statistical calculations is a dimensionless quantity. For the first n The temperature anomaly observed is in °C. This represents the average value of temperature anomalies. For the first n The sea surface temperature anomaly observed this time is measured in °C. This represents the average value of sea surface temperature anomalies.

[0084] This is a deep background response, expressed in °C / °C. , In the first n In this observation, at a selected deep reference depth ( Temperature anomalies measured or calculated at the location The depth should be selected at a depth where the effects of turbulent mixing have largely disappeared and large-scale processes dominate.

[0085] The attenuation depth scale is in meters. , ρ is the turbulent diffusion coefficient, which physically quantifies the efficiency of vertical mixing of heat and momentum caused by turbulence (eddies) in the ocean. Its unit is m² / s. A higher value indicates stronger turbulent mixing and higher efficiency in downward heat transfer. .

[0086] Friction velocity is a speed measure of the momentum flux (i.e., wind stress) at the air-sea interface, measured in m / s. Stronger winds result in greater wind stress. The larger the value, the more it is determined by sea surface wind stress. and seawater density Decide: .

[0087] h Boundary layer depth, usually referring to the depth of the ocean mixing layer or boundary layer, is measured in meters. It is the depth of the upper ocean layer that is fully mixed and homogeneous by processes such as wind and waves, and is the main area affected by turbulent mixing.

[0088] The Richardson number is a dimensionless number and a key parameter for judging fluid stability. It compares the stabilizing factors that suppress turbulence (based on density gradient). (representative) and factors that promote instability in turbulence (by velocity shear) (Representative), its calculation formula is as follows , This indicates stable flow and that stratification suppresses turbulent mixing. This indicates unstable flow and promotes the development of turbulence. This represents neutral stratification; in this model, larger positive values ​​indicate... The number will significantly suppress the turbulent diffusion coefficient Thus, the attenuation depth scale It gets smaller. (In Richardson's number) In the context of the formula, Specifically, it refers to the vertical gradient of density, that is, the rate and direction of density change in the vertical direction (depth direction), and its complete mathematical expression is: , Represents partial derivatives. In Richardson's number... In the context of the formula, Specifically refers to the rate at which horizontal velocity changes in the vertical direction (depth direction), and its complete mathematical expression is: .

[0089] (2) .

[0090] in, The climatological temperature gradient is a background field representing the degree of temperature variation in the ocean itself in the vertical direction, expressed in °C / m. , The climatological temperature gradient is a continuous form and is a mathematical partial derivative. For the approximate calculation of discrete climatological temperature gradients, The climatological average temperature at depth z represents the vertical temperature distribution of a specific sea area in a long-term statistical sense. The climatological temperature difference is the climatological mean temperature change within the depth interval [z-Δz, z+Δz]. For depth intervals.

[0091] The isotherm vertical displacement efficiency factor is a dimensionless proportionality coefficient, which represents the efficiency of vertical displacement at depth. z The vertical displacement of isotherms caused by a unit sea level height anomaly (SLA) at a given location. η ).Right now , For depth z The vertical displacement of the isotherm at a given location, in meters.

[0092] The attenuation scale depth is expressed in meters. The larger the value, the deeper the influence of the dynamic process can reach, which is usually related to the dynamic processes of positive pressure and the first inclined pressure die.

[0093] This represents the attenuation of the first mode (or main mode), indicating that the influence of isothermal displacement decreases exponentially with depth starting from the sea surface.

[0094] C It is an amplitude constant, dimensionless, used to control the intensity of contributions from higher-order modes.

[0095] The depth at which the higher-order modes have the greatest influence is measured in meters. At this depth, the isothermal displacement exhibits the strongest response to SLA.

[0096] The width of the Gaussian function, in meters, determines the range (thickness) of the influence of the maximum value. The larger it is, the wider its influence.

[0097] (3) Based on the heat budget equation of the mixed layer: ;in, The time trend term for the temperature anomaly of the mixed layer represents the rate of change of the temperature anomaly of the mixed layer over time, i.e. the rate of heating or cooling, in °C / s. This is the vertical temperature gradient at the bottom of the mixing layer (the temperature gradient of the thermocline not far below the bottom of the mixing layer), typically (Temperature decreases with increasing depth). This represents the net heat flux over the sea surface. The heat entering the ocean is positive (heating the ocean), and the heat leaving the ocean is negative (cooling the ocean). The specific heat capacity of seawater. For the depth of the mixing layer, For winding speed, , k This is a proportionality constant, which is usually a positive number.

[0098] The final form is a piecewise Gaussian function: .

[0099] in, It is a nonlinear intensity. and The center depth of the Gaussian function represents the depth at which the temperature anomaly response caused by the entrainment process is greatest. Corresponding to cold vortex, Corresponding to a warm vortex, the unit is meters. and The width (standard deviation) of the Gaussian function controls the extent of the temperature anomaly response in the vertical direction. Corresponding to cold vortex, Corresponding to a warm vortex, the unit is meters.

[0100] (4) The displacement efficiency function is extended to: .

[0101] Substitution expression: .

[0102] The final form is: .

[0103] in, The isothermal displacement efficiency factor (extended version) depends on depth and sea surface temperature anomalies, representing the energy output per unit SLA at a specific depth under specific SST' conditions. z The magnitude of the isothermal displacement caused at a certain point is dimensionless. The basic displacement efficiency function represents the thermocline response without SST disturbance. Let be the modulation coefficient function, representing the depth z At this location, the unit SST' anomaly can affect the displacement efficiency. The percentage change is expressed in units of 1 / °C. The peak amplitude of the coupling strength SST' modulation intensity is a constant in °C. -1 . The maximum coupling depth (the core depth of the thermocline) is a constant in m. The coupling thickness is a constant in meters.

[0104] This application uses gradient descent to solve for each regression coefficient, and the steps are as follows.

[0105] (1) Initialize each regression coefficient , b=1,2,3,4 This represents the initial guess value of the regression coefficient, and the superscript (0) indicates the 0th iteration.

[0106] (2) Iterative update: ;in, Step size This means that each update only makes very small adjustments to the coefficients, which helps the optimization process to converge stably to the minimum point, rather than oscillating around the minimum or even diverging. For the first The regression coefficients after the next iteration. For the first The regression coefficients are updated after the next iteration. MSE, or mean squared error, measures the predicted temperature. Compared with observed values The smaller the average difference between the two models, the better the model fits. Mean squared error versus regression coefficients The partial derivative of indicates the degree of influence of the MSE function at the current coefficient values. The steepest upward direction in terms of direction.

[0107] (3) Regularization: Introduce an L2 penalty term (hyperparameter λ=0.01) to constrain the magnitude of the coefficient to prevent overfitting.

[0108] (4) Convergence criterion: relative error change or number of iterations q >10 4 If the first criterion is not met after 10,000 iterations, the algorithm is forced to stop to prevent it from getting stuck in an infinite loop or wasting computational resources due to slow convergence.

[0109] The objective function optimized by the least squares method is: . This is an L2 penalty term used to penalize excessively large coefficient values. The regularization strength hyperparameter controls the proportion of the penalty term in the overall objective function. This means that the algorithm still primarily aims to minimize the fitting error, while gently constraining the size of the coefficients to prevent them from becoming absurdly large.

[0110] Step 205: Based on the regression coefficients, the mesoscale eddy three-dimensional temperature field inversion model is used to invert the mesoscale eddy three-dimensional temperature field. At the same time, dynamic constraints are embedded in the inversion process to obtain the temperature of each standard layer of the mesoscale eddy.

[0111] The dynamic constraints include quasi-geostrophic equilibrium constraints and sea surface boundary conditions.

[0112] Quasi-geostrophic equilibrium constraint: Based on the quasi-geostrophic characteristics of mesoscale eddies, a closed-loop relationship between temperature, pressure, and flow velocity is established.

[0113] Pressure-temperature correlation: .

[0114] Earth rotation velocity: , .

[0115] Sea surface boundary conditions: .

[0116] Among them, the integration path starts from depth z To the sea ( z =0), which reflects the vertical integral characteristic of static equilibrium. For radial distance r ,depth z Pressure anomalies at a location refer to the deviation of pressure from the climatological mean field caused by ocean eddies and other activities, and are measured in Pascals (Pa) or decibars (dbars). It's important to note that ocean surface pressure is usually assumed to be constant (atmospheric pressure), so the pressure here... It actually refers to the pressure change acting on the seabed barometer caused by abnormal density distribution within the ocean, rather than a true atmospheric pressure anomaly.

[0117] ρ is the coefficient of thermal expansion, representing the relative rate of change of seawater volume when the temperature increases by 1°C under constant salinity and pressure. For typical seawater, It is a positive value (approximately 2 × 10⁻⁶). -4 / °C) means that increasing the temperature will cause the density to decrease (volume expansion).

[0118] The reference density of seawater is taken as a constant value (e.g., ~1025 kg / m³), meaning that the formula ignores the feedback of density anomalies on density itself. For radial distance r ,depth z Temperature anomalies at the location, measured in degrees Celsius (°C).

[0119] This represents the eastward (x-direction) component of the geostrophic flow velocity. This represents the northward (y-direction) component of the geostrophic velocity, measured in meters per second (m / s). The sign of the geostrophic velocity indicates the direction of the Coriolis force; the flow direction always causes the Coriolis force to point to the left of the flow direction (in the Northern Hemisphere) to balance the pressure gradient force from high pressure to low pressure. Pressure anomaly p 'Gradient in the horizontal direction,' This represents the rate of change of pressure anomalies in the eastward direction, expressed in Pascals per meter (Pa / m). Geostrophic equilibrium requires flow along isobars; in the Northern Hemisphere: if pressure increases in the y-direction (…), then… This will produce a flow velocity along the -x direction ( If the pressure increases in the x-direction ( This will generate a flow velocity along the +y direction ( ).

[0120] Further vertical mode separation was performed: Pressure Anomaly Decomposition: Its core assumption is that a three-dimensional pressure anomaly field of a mesoscale vortex (or other quasi-geostrophic motion) can be decomposed into a field that is independent of the radial distance. r Related functions and only with depth z Related functions The product of these vortices, i.e., the horizontal structure of vortices at all depths is "self-similar," except that the amplitude varies with depth. The horizontal structure function of the pressure anomaly describes how the pressure anomaly of the vortex is distributed and attenuated in the horizontal direction. For example, it determines the radius and central intensity of the vortex. The vertical structure function of the pressure anomaly, also known as the vertical mode, describes how the amplitude of the pressure anomaly of the vortex changes with depth, and also determines the vertical structure of the flow velocity, etc.

[0121] Radial equation: ,untie: That is, the modified Bessel function. Wherein, Let the radius of deformation be the radius of the Rossby deformation. For radius, For the second type of modified Bessel function (order 0), the vortex core structure, in The limited divergence reflects the intensity at the vortex center and the attenuation characteristics at the periphery of the vortex. The exponential attenuation indicates that the vortex's influence range is limited.

[0122] Vertical equation (non-homogeneous boundary): .in, The value is the square of the buoyancy frequency, a key parameter for measuring the stability of ocean stratification. The unit is squared per second. The larger the value, the stronger the ocean stratification and the more stable the stratification, representing the restoring force of buoyancy to disturbances in the vertical direction. The deformation radius, or Rossby radius, is one of the most important scale parameters in stratified fluids on rotating Earth, representing the equilibrium scale between rotational and stratification effects. The horizontal scale of motions (such as vortices and planetary waves) is typically... They have the same order of magnitude. The larger the value, the stronger the stratification effect, and the wider the horizontal range that the movement can influence. , To reduce gravity, , H represents the density difference, and H is the depth of the entire water layer or the equivalent depth of the stratified layer.

[0123] Boundary condition: G|z= p 0, .

[0124] β2 physical constraints: .in, The intensity of the forcing applied to the ocean surface (z=0), i.e., the intensity of the surface forcing, is the external source driving the fluid motion (such as surface pressure anomalies or heat flux), and its value is fixed by the boundary conditions. p 0.

[0125] The effectiveness of the three-dimensional temperature field inversion method for ocean mesoscale eddies provided in this application will be verified below.

[0126] (1) Objective analysis of interpolation.

[0127] Outlier removal is achieved by using interquartile range (|deviation|>3×interquartile range). Interquartile range is a statistic that measures the dispersion of data. It is calculated as the difference between the 75th percentile (Q3) and the 25th percentile (Q1) of the data, representing the distribution range of the middle 50% of the data, and is not sensitive to extreme values.

[0128] Gaussian correlation function weighted interpolation: , .in, Represents target grid points The temperature value at that location. To give the first n The weight of the observation point in each observation depends on the spatial distance between the observation point and the target grid point. For the first n The actual temperature (anomaly) measurement value of the observation point for this observation. For the first n The radial distance between the observation point and the target grid point in this observation. The correlation scale is a key parameter in the Gaussian function, which determines how quickly the weights decay with distance.

[0129] (2) Composite analysis: Establish a polar coordinate system with the vortex center as the origin, and follow the... r / R Boxed (0.1) R (Interval), calculate the median value of T′ in each box.R Let be the radius of the vortex. r / R This is a dimensionless radial distance, done to normalize vortices of different sizes, allowing their structures to be compared and synthesized. 0.1 R The interval indicates the width of the compartments, which will cause the vortex to move from the center ( r =0) to the periphery (e.g. r =2 R The data is divided into a series of continuous annular regions (“boxes”) with a width of 0.1 times the vortex radius. The median of T' represents the median of all temperature anomalies within each annular box. The median is used instead of the mean because the median is not sensitive to outliers or asymmetric distributions that may remain within the box and better describes the typical characteristics of the data.

[0130] Radial smoothing: Savitzky-Golay filter (window=5, order=2), formula: .in, For radial distance r At that point, the smoothed temperature anomaly value, m For the index of data points within the window, m =0 corresponds to the center point. m =-1 and m =-2 corresponds to the point on the left. m =1 and m =2 corresponds to the point on the right. The convolution coefficients of the Savitzky-Golay filter are uniquely calculated and determined by the specified window size and polynomial order. Essentially, they are the weights used to perform a weighted average of the data within the window. The design of these weights ensures that the smoothed curve can retain the characteristics of the original data's trend (such as peaks and valleys) to the greatest extent possible. The horizontal spacing is the radial distance between the centers of two adjacent compartments.

[0131] (3) Algorithm Validation: To verify the reliability of the algorithm and the accuracy of the algorithm's inversion results, all data from 2018 were selected for comparison (0-60°N, 0-180°E in 2018). Actual sea surface temperature observation data (Argo data from 2018) were used as evaluation data and compared with the inverted temperatures within the same latitude and longitude range at the same time. Selecting all data from one year was to ensure that the dataset covers different time periods and the corresponding ocean temperature conditions for different time periods, and to conduct a comprehensive evaluation.

[0132] BIAS measures the deviation between the inversion results and the true values ​​(2018 Argo data). The formula is: ;in, The temperature data for 2018 are obtained after inversion (0-60°N, 0-180°E). This is Argo observation data from 2018.

[0133] The correlation coefficients and error rates between the automatically extracted vortex equivalent radius and vortex velocity and the corresponding physical quantities from the 2018 AVISO mesoscale vortex data were calculated, and the correlation coefficient R1 and error rate were also determined. Error Rate : ; , for The mean, for The mean.

[0134] This application addresses the practical need for high accuracy and real-time performance in the inversion of the three-dimensional structure of mesoscale eddies. Based on field, historical, and reanalysis data, it utilizes regression analysis to construct a targeted three-dimensional temperature field inversion model for cold / warm mesoscale eddies. By comprehensively utilizing field observations and satellite remote sensing, it achieves real-time and accurate inversion of the vertical structure of mesoscale eddies and conducts field observation experiments for verification, providing data support for maritime activities. Specifically, the OW method is used to identify the edges of mesoscale eddies at the sea surface, and the WA method is used to identify eddy edges through flow fields in the 5m-200m layer below the sea surface. The three-dimensional temperature inversion of eddies employs regression analysis, constructing a mathematical model based on various historical data to achieve the inversion of eddy temperatures. The historical data used encompasses multiple data sources, including objective reanalysis data, Argo historical observation data, and satellite data. Through in-depth mining and integration analysis of these rich and representative data resources, a mathematical model is constructed that can accurately characterize the relationship between the three-dimensional temperature field of the ocean and sea surface temperature, sea level height, and sea level anomalies. This model can accurately predict the three-dimensional temperature distribution of each level of eddies. The results show that the constructed mesoscale eddy three-dimensional temperature field inversion model exhibits high accuracy in multiple validation cases, with correlation coefficients above 0.9 and mean square errors below 0.5°C.

[0135] Based on the same inventive concept, this application also provides a three-dimensional temperature field inversion device for ocean mesoscale eddies to implement the above-mentioned method for inverting the three-dimensional temperature field of ocean mesoscale eddies. The solution provided by this device is similar to the solution described in the above-described method. Therefore, the specific limitations of one or more embodiments of the three-dimensional temperature field inversion device for ocean mesoscale eddies provided below can be found in the limitations of the three-dimensional temperature field inversion method for ocean mesoscale eddies described above, and will not be repeated here.

[0136] In one exemplary embodiment, such as Figure 3As shown, a three-dimensional temperature field inversion device for ocean mesoscale eddies is provided, comprising: a data acquisition module 301, a mesoscale eddy identification module 302, a model construction module 303, a coefficient solving module 304, and a temperature inversion module 305.

[0137] The data acquisition module 301 is used to acquire multi-source ocean observation data.

[0138] The mesoscale eddy identification module 302 is used to identify mesoscale eddies based on the multi-source ocean observation data and determine the three-dimensional structure of the mesoscale eddies.

[0139] The model building module 303 is used to construct a three-dimensional temperature field inversion model of a mesoscale eddy based on the multi-source ocean observation data and the three-dimensional structure of the mesoscale eddy, using regression analysis.

[0140] The coefficient solving module 304 is used to solve the regression coefficients in the mesoscale eddy three-dimensional temperature field inversion model using the hierarchical least squares method.

[0141] The temperature inversion module 305 is used to invert the three-dimensional temperature field of the mesoscale eddy using the mesoscale eddy three-dimensional temperature field inversion model based on the regression coefficients, and to embed dynamic constraints during the inversion process to obtain the temperature of each standard layer of the mesoscale eddy.

[0142] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 4 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs in the non-volatile storage media to run. The database stores multi-source ocean observation data. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When executed by the processor, the computer program implements a method for inverting the three-dimensional temperature field of oceanic mesoscale eddies.

[0143] Those skilled in the art will understand that Figure 4The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0144] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0145] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0146] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0147] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0148] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0149] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0150] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for inverting the three-dimensional temperature field of ocean mesoscale eddies, characterized in that, The method includes: Acquire multi-source ocean observation data; Based on the multi-source ocean observation data, mesoscale eddies are identified, and their three-dimensional structures are determined. Based on the multi-source ocean observation data, and according to the three-dimensional structure of the mesoscale eddy, a three-dimensional temperature field inversion model of the mesoscale eddy is constructed using regression analysis. The regression coefficients in the mesoscale eddy three-dimensional temperature field inversion model are solved using the hierarchical least squares method. Based on the regression coefficients, the mesoscale eddy three-dimensional temperature field inversion model is used to invert the mesoscale eddy three-dimensional temperature field, while embedding dynamic constraints during the inversion process to obtain the temperature of each standard layer of the mesoscale eddy. The multi-source ocean observation data includes buoy temperature and salinity datasets, satellite altimeter data, ocean climate datasets, and drifting air-sea interface buoy observation data; Based on the aforementioned multi-source ocean observation data and the aforementioned three-dimensional structure of mesoscale eddies, a three-dimensional temperature field inversion model for mesoscale eddies is constructed using regression analysis, including: Determine sea surface height anomaly data based on the satellite altimeter data; Based on the buoy temperature and salinity dataset, the ocean climate dataset, and the observation data from the drifting air-sea interface buoy, sea surface thermal anomaly data were determined. Based on the sea surface thermal anomaly data and the sea surface height anomaly data, determine the thermodynamic coupling data; Using the sea surface height anomaly data, the sea surface thermal anomaly data, and the thermodynamic coupling data as independent variables, and the temperature of each standard layer of the mesoscale eddy as the dependent variable, a three-dimensional temperature field inversion model of the mesoscale eddy is constructed by regression analysis.

2. The method for inverting the three-dimensional temperature field of ocean mesoscale eddies according to claim 1, characterized in that, Based on the aforementioned multi-source ocean observation data, mesoscale eddies are identified, and their three-dimensional structure is determined, including: For the sea surface area, based on the satellite altimeter data, the parametric method is used to identify mesoscale eddies in order to determine the three-dimensional structure of mesoscale eddies in the sea surface area; For the subsurface region, based on the satellite altimeter data, the winding angle method is used to identify mesoscale eddies in order to determine the three-dimensional structure of mesoscale eddies in the subsurface region.

3. The method for inverting the three-dimensional temperature field of ocean mesoscale eddies according to claim 1, characterized in that, The thermodynamic coupling data is the product of the sea surface thermal anomaly data and the sea surface height anomaly data.

4. The method for inverting the three-dimensional temperature field of ocean mesoscale eddies according to claim 2, characterized in that, The inversion model for the three-dimensional temperature field of the mesoscale eddy is as follows: ； in, For depth Temperature at the standard floor level This is data on sea surface thermal anomalies. This is data on sea level height anomalies. For thermodynamic coupling data, For random error term, The efficiency of vertical transfer of sea surface thermal anomalies to deep water. , , , For regression coefficients, The linear response coefficient representing the sea surface thermal anomaly. The linear response coefficient representing sea level height anomalies, The nonlinear response coefficient representing the sea surface thermal anomaly. This represents the thermodynamic coupling response coefficient.

5. The method for inverting the three-dimensional temperature field of ocean mesoscale eddies according to claim 1, characterized in that, The dynamic constraints include quasi-geostrophic equilibrium constraints and sea surface boundary conditions.

6. A device for inverting the three-dimensional temperature field of ocean mesoscale eddies, characterized in that, The apparatus is applied to the three-dimensional temperature field inversion method for ocean mesoscale eddies as described in any one of claims 1-5, and the apparatus comprises: The data acquisition module is used to acquire multi-source ocean observation data; The mesoscale eddy identification module is used to identify mesoscale eddies based on the multi-source ocean observation data and determine the three-dimensional structure of the mesoscale eddies. The model building module is used to construct a three-dimensional temperature field inversion model of a mesoscale eddy based on the multi-source ocean observation data and the three-dimensional structure of the mesoscale eddy using regression analysis. The coefficient solving module is used to solve the regression coefficients in the mesoscale eddy three-dimensional temperature field inversion model using the hierarchical least squares method. The temperature inversion module is used to invert the three-dimensional temperature field of the mesoscale eddy using the mesoscale eddy three-dimensional temperature field inversion model based on the regression coefficients, and to embed dynamic constraints during the inversion process to obtain the temperature of each standard layer of the mesoscale eddy.

7. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the method for inverting the three-dimensional temperature field of ocean mesoscale eddies according to any one of claims 1-5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the method for inverting the three-dimensional temperature field of ocean mesoscale eddies as described in any one of claims 1-5.

Images